The impact of the national antimicrobial stewardship programmes on clinical outcomes: a baseline trend analysis

Problem

The increasing use and sometimes unnecessary consumption of antibiotics is a key driver of antimicrobial resistance. A range of national initiatives was introduced to reduce antibiotic prescribing, for example, the “Quality Premium: 2015/16 guidance” which included financial incentive to help reduce unnecessary antibiotic prescribing in primary care. In order to examine any unintended consequences of this policy, it is important to establish baseline trends in infection.

Approach

We used the national Hospital Episode Statistics (HES) dataset from April 2010 to March 2015 to assess admissions relating to pre-defined potential unintended complications from a reduction in antibiotic prescribing in respiratory, urinary tract, and other clinical infections syndromes. We used direct standardisation method to estimate age-standardised rates per 100,000 population and adjusted for seasonal variation. We graphically compared unadjusted and seasonally adjusted time series for all infections as a single group and, where possible, individually. We obtained national community prescribing data from Information Service Portal from October 2011 onwards.

Findings

We identified over 3.1 million emergency hospital admissions during the study period. The age-standardised hospital admission rate increased from 1,032 per 100,000 population in 2010/11 to 1,271 per 100,000 population in 2014/15. The hospital admission rate increased for almost all infections. There was a marked increase in the age-standardised hospital admission rates for sepsis from 46 per 100,000 population in 2010/11 to 92 per 100,000 in 2014/15. The hospital admission rate for scarlet fever increased from 0.9 per 100,000 to 1.9 per 100,000 in the same period. Identifiable seasonality was present in all infections, excluding empyema, brain abscess, rheumatic fever and sepsis. Hospitalisation for community-acquired and hospital-acquired pneumonia was highest during the winter period, while hospitalisation for pyelonephritis reached a peak in late summer/early autumn. Quarterly data for overall and broad-spectrum (co-amoxiclav, cephalosporins and quinolones) antibiotic prescribing showed seasonal variation with a higher number of antibiotic prescriptions in winter months compared with summer months. Comparing the financial year 2014/15 (April 2014 – March 2015) to the financial year 2013/14, an increase in overall antibiotic prescriptions was detected; however, the broad-spectrum antibiotic items dropped by 3.4% in the financial year 2014/15 compared financial year 2013/14.

Consequences

These trends will be used as a baseline to determine the impact of antimicrobial stewardship programmes as more recent data become available. An interrupted time series design and segmented regression analysis will be used to evaluate the impact of the programme on clinical outcomes.